Changing the number of retrieved WebShop trajectories from k=3 to k=5 changes success rate by less than two points for every executor; the differences are within standard deviation for Qwen3-8B and GPT-5.4, and the main experiments use k=3.

Changing the number of retrieved WebShop trajectories from k=3 to k=5 changes success rate by less than two points for every executor; the differences are within standard deviation for Qwen3-8B and GPT-5.4, and the main experiments use k=3. · CiteArk

测量结果

已评估 0/12 项

获得支持 0有争议或冲突 0尚未评估 12

保留同一结论下的完整结果组。已有评估排在前面；单项判断不替代整组结论。展开行查看条件、来源与历次证据。

每行是一项测量，分别显示论文报告、最近可用实测和累计证据状态。
论文报告	最近可用实测	证据判断
61.1 points	—	尚未评估
32.8 percentage points	—	尚未评估
60.5 points	—	尚未评估
32.8 percentage points	—	尚未评估
61 points	—	尚未评估
50.5 percentage points	—	尚未评估
59.2 points	—	尚未评估
48.6 percentage points	—	尚未评估
53.8 points	—	尚未评估
45.4 percentage points	—	尚未评估
53.4 points	—	尚未评估
45.3 percentage points	—	尚未评估

Changing the number of retrieved WebShop trajectories from k=3 to k=5 changes success rate by less than two points for every executor; the differences are within standard deviation for Qwen3-8B and GPT-5.4, and the main experiments use k=3.

报告指标与观测值

Assessment（0）

关联运行（0）

测量结果